LATE HOMO ERECTUS OR NGANDONG MAN OF JAVA

G.-J. Bartstra Biologisch-Archaeologisch InstitulII, Groningen, Netherlands

ABSTRACT: Ngandong man of Java is generally regarded as the last representative of Homo erectus in Southeast Asia. His remains came from terrace deposits of the Solo river. This paper investigates the age of these deposits, as well as the evidence of stone tools associated with Ngandong man.

KEYWORDS: Homo erectus, Ngandong man, Java, Solo river ten:aces, vertebrate fossils, U-series dating, artifacts.

1. INTRODUCTION in eas tern Java, contains cultural remains that are not older than Lower Holocene or Final Plelstocene Since 1970 I have had the opportunity to take part (van Heekeren, 1972; Hardjasasmita & Mulyana, in research on fossil man of Java. I am especially 1986). interested in the distribution of the Palaeolithic It may have something to do with fire. It may be artifacts and the terrace systems of the ·rivers. The argued that early man only occupied the tropicai two subjects are c1 0sely connected, for the study of caves permanently when he had fully mastered the the implements of fossil man on Java is impossible use of fire. Homo erectus on Java, even late Homo without detail ed geomorphological surveys. To use erectus, perhaps still did not know how to effec­ a term that was much en vogue at the beginning of tively handle fire. Nowhere on Java is there any this century: on Java one is concerned with so­ direct evidence of the use of fire by Homo erectus. It called river-drift Palaeolithic. The majority of the has merely been inferred, from evidence from other Palaeolithic artifacts are to be found in fluviatile Homo erectus sites elsewhere, that on Java' too, sediments. There is little chance of finding un­ Homo erectus must have been able to handle fire disturbed camp-sites dating from the Pleistocene. (Pope, 1985). But the romantic view of the band of Any former coastal sites now lie under water as a Homo erectus people sitting around that fire, over result of the post-Pleistocene marine ingression, which the kill of the big game hunt is being while farther inland layers of volcanic ash and prepared, has come under attack (e.g. Binford & mudstreams, up to several metres thick, cover the Stone, 1986). It has also been pointed out that the land. The caves appear to be devoid of traces of dentition of Homo erectus of Java shows traces of early man. wear resulting from a diet mainly consisting of When Dubois, the dis coverer of the first Homo coarse, tough vegetable matter (Puech, 1983). erectus, arrived in the East in 1887, he began to It is clear that the analysis of the tools used by excavate caves on the island of Sumatra. But the Homo erectus on Java could shed light on be­ results were disappointing: remains of early man havioural questions of this kind. Unfortunately still were absent (e.g. Hooijer, 1948a; 1948b; 1949). far toa little is known about these tools. Stone There are caves on Java too, and many ofthem have artifacts are to be found all over Java, but only a now been investigated, but early Pleistocene man is small percentage of them can be attributed with not to be found there. Von Koenigswald (1955), certainty to Homo erectus (Bartstra, 1982; Bartstra who discovered the second Homo erectus on Java, & Basoeki, in press). Concerning bone artifacts provided a psychological explanation: early Plei­ nothing is certain at all. stocene man avoided the tropicai caves because they With regard to implements and possibie beha­ were unhealthy places, with snakes, bats, and eviI viour, at the moment we know most about the last spirits. However, this does not explain why later representative of Homo erectus on Java: Ngandong prehistoric man, fully developed Homo sapiens, did or Solo man. During the·last few years my own dwell in caves, on both Sumatra and Java. Gua fieldwork on Java has concentrated on the.evidence Lawa (Sampung), for example, a well-known cave of this hominid in particular. 2 G.-J. BARTSTRA

D J A V A S E A

Ngondong J A VA •

'-Trlnil 'lAWUVOlUIIO

• Sompung

=

I N D I A N o C E A N Fig. l. Map of eastern Java, showing localities referred to in the text.

2. THE DISCOVERY OF NGANDONG MAN Teilhard de Chardin (1956). Von Koenigswald (1958) stood in faet quite alone in describing Ngandong is a small village tucked away in the Ngandong man as a tropicai Neanderthal variety. forests of teak (Teetona grandis) that cover the In the 1930's an interesting discussion started Kendeng hills, a low anticlinal ridge in Central about the supposed eannibalism ofNgandong man. Java. The village is situated close to the Solo river, Eye-witnesses of the original excavations stat ed that which originates in the mountains of southern Java, the skulis lay embedded in the river sediments in a subsequently flowing eastwards to the south of the rather particular way, in a position in whieh they Kendeng ridge, then breaking through these hills, Offered preeisely the greatest resistance to the and finally debouching into the Strait of Madura former stream current: with the concave basal side (fig. 1). upwards (ter Haar, 1934). In addition a few of the It was in the 1930's that the surroundings of calvaria show remarkable fractures: the area around Ngandong were first mapped geologically and geo­ the lommen magnum is either damaged or even graphically in detail. The mining engineer ter Haar missing (von Koenigswald, 1951). It has therefore was entrusted with this task, and during surveys been assumed that Ngandong man was a cannibal along the Solo banks he found vertebrate fossils. who ate members of his own species, and that the Ter Haar started a small excavation, that was soon skulis had been intentionally left behind in a special taken over and extended by Oppenoorth, at that way; or, alternatively, a more highly developed time head' of the Java mapping project. During hominid, who has not yet been found at Ngandong, these excavations, that lasted from 1931 until 1933, " may have preyed upon this Ngandong man. No­ a wealth of vertebrate remains were found in old tably von Koenigswald (1958) was a proponent of river sediment, among them eleven calvariae and the cannibalistic vie w . I myself have stated that the calottes and two tibiae of early hominids. arguments in favour of this view' appear rather Oppenoorth (1932a) reported the discovery as convincing (Bartstra, in press a). Others, however, Homo (Javanthropus) soloensis; ter Haar (1934) later remain sceptical (Jacob, 1972; Santa Luca, 1980). wrote an interesting paper about the circumstances Yet it is clear that if the use of fire and the practice of of the find. Subsequently others devoted their big game hunting by Homo erectus on Java come attention to Ngandong or Solo man, e.g. Weiden­ under discussion, new light has to be shed also on reich (1951), von Koenigswald (1958), Jacob (1978), this supposed cannibalistie behaviour. Sartono (1980), and Santa Luca (1980). Santa Luca describes Ngandong man as a late variety of Homo erectus on Java, with robust supraorbital, angular 3. THE TERRACES OF THE SOLO RIVER and occipital tori. The average cranial capacity was 1100-1200 cc. The tibiae appear modern in form (see Oppenoorth (1932b) pointed out aiready that Ngan­ also Day, 1986). The idea that Ngandong man has dong man must be younger than the other varieties strong affinities with earlier Homo erectus (Pithe­ of Homo erectus on Java. This is suggested by not canthropus erectus) on Java, had aiready been only morphological but also geological evidence. stressed by e.g. Weidenreich (1945; 1951) and The calvariae and tibiae from Ngandong come from Late Homo erectus ar Ngandong man af Java 3 sandy, gravelly stream sediments that show a clear coarser than those of the Low terrace; the latter is association with the drainage pattern of the Solo known as the 'clay 'terrace', as the basal con­ river. EarIier Homo erecfus of Java (TriniI or Java glomerate (if present) lies mostly buried under man) comes from stream sediments that cannot be several metres of silt or clay. connected with the present-day Solo. The fossil­ After Lehmann, de Terra (1943) investigated the bearing sediments near Ngandong are so-called Solo terraces. He distinguished three leveIs, and in ter race deposits,· or fossil remnants of a former some places even more. Yet if one tries to trace de floodplain, that are now to be found high above the Terra's levels in the field one meets with difficulties. present stream. One cannot help thinking that de Terra tried to fit The Solo terraces have been a subject of discus­ real and imaginary Solo terrace levels into a system sion since the beginning of this century. InitiaIly of four glaciais, that was customarily accepted in most researchers refused to accept that terrace the 1940's, following the European model. But deposits even existed; in this respect the German breaks in the valley slopes that de Terra described as geologist Elbert (1908) was an exception. But then terrace scarps are often impossible to trace in the ter Haar found vertebrate fossils in stream sedi­ field (Bartstra, 1977). ments lying at least 20 m above the present-day Sartono (1976) has mentioned seven Solo terra­ river. Moreover, these sediments contain small ces. He established the presence of these on the basis pebbles of volcanic rock that do not occur in the of evidence provided by aerial photography. How­ wide surroundings of Ngandong, and that must ever, the two oldest terraces and the youngest have been brought by the Solo from farther south. terrace that he distinguished are not true Solo Thus, in the 1930's it was definitely acknowledged ter'races. The two oldest ones have no association that terrace deposits exist. with the drainage pattern of the present-day Solo; Some authors only mention stream terraces in the they date from a time of reversed drainage, when Solo valley directly to the north and south of streams flowed southwards from the Kendeng hills. Ngandong, where the Solo river cuts through the Sartono's youngest terrace can better be interpreted Kendeng hills. But terrace deposits are also present as a modern floodplain feature: low accumulations beyond these hills, where the Solo meanders into a of cobbles and boulders reminiscent of recent lowland region. There, the old fills are buried under floods. In fact Sartono thus distinguishes four Solo the youngest stream-gravels and sands (fig. 2). This terraces, and in this respect does not differ much fill-in-fill phenomenon differs markedly from the from de Terra. classic step-like terrace flights near Ngandong. In my opinion for the time being it is better to Lehmann (1936) gave attention to the Solo speak of only a High and a Low terraCe along the terraces with regard to geomorphology. He distin­ Soio. A lot of field research will yet have to be done guished two terrace leveis, that he called the High to find out whether the faint breaks that can be and the Low terrace, situated 20 m and 5 m above traced here and there on the valley slopes can really the river. As is to be expected in a tropicaI climate be interpreted as former terrace scarps. Because of with severe erosion, the High terrace is dissected. It tilting and warping one cannot rely on height has only remained preserved in a few sheltered measurements alone for the interpretation bf ter­ places, like that found near the village ofNgandong. race leveIs, and sedimentological analyses are necesc The constituents of the High terrace are distinctly sary.

N s

::-: :.:.: .... :-: ' ...... � 1 o Skm 1:.:::::,. .. .- ...- ...... �2

Fig. 2. Cross-section north of Ngandong, showing the fill-in-fill phenomenon: I. High terraee fill; 2. volcanic bedrock; 3. limestone bed rock; 4. Low terrace fill (modified after van Bemmelen, 1949). 4 G.-J. BARTSTRA

4. THE VERTEBRATE FOSSILS OF Ngandong man or Homo erectus soloensis In the NGANDONG modern nomenc1ature. This age has not been dis­ puted by most palaeontologists. The first reports of the fossils of Ngandong were An exception is Santa Luca (1980), who is of the rapturous. Oppenoorth (1932a; 1932b) acknow­ opinion that the hominid remains from Ngandong ledged the extraordinary richness of the fossil­ could have been reworked, and that they might not bearing terrace fill, and von Koenigswald (1933), be contemporaneous with the rest of the fauna. who regularly visited Ngandong during the years of However, researchers who were present at the the excavations, referred to it as one of the richest original excavations in the 1930's have never ex­ and most important sites on Java. He also indicated pressed any doubts about the uniformity of the that complete vertebral columns and skulis with material found. AIso Jacob (1975) has stated that lower jaws still attached had been found, which is on the basis of chemical analyses there seems to be not indicative of much reworking of the ani mal no difference between the hominid cranial remains remains (von Koenigswald, 195 1). An interesting and the ani mal fossils. Santa Luca's suggestion has feature of the Ngandong vertebrate list is the large probably been prompted by concern about the find number of bovids represented. The horn span of e.g. of a Homo erectus species in relatively young, Upper Bubalus palaeokerabau is suggestive of a fauna that Pleistocene deposits. was adapted to an open forest environment, as was The question leads to taphonomic considera­ aiready indicated by ter Haar (1934), and recently tions. The Kendeng hills near Ngandong consist of again by de Vos (1985). It would be interesting to limestone and maris; only the terrace deposits provide support for these views by means of pollen contain vo1canic constituents from farther south. In research, but a few samples of terrace fill collected principle it is possibie that from the older deposits near Ngandong in 1986 turned out to be devoid of eroded by the Solo in the south (the so-called pollen grains. Notopuro and Kabuh layers) fossil fragments have Von Koenigswald was employed in the Geo­ been transported downstream, that consequently in logical Service of the former Dutch East Indies as a the High terrace must be regarded as allochthonous vertebrate palaeontologist. When he arrived on components. But the distance from Ngandong to Java, in 193 1, it was cu�tomary to make an estimate the nearest outcrops of these older deposits is at of the age of Tertiary cif{(jQuaternary sediments on least 30 km (measured along the meandering river). the basis of the in situ invertebrate fauna. In making Such a distance is acceptable for small, aiready such estimates the percentage of recent, still extant fossilized fragments, carried along with the channel­ species, as well as of fossil species, is of importance load, but not for eleven hominid skulis. One must (Martin, 19 19; van der Vlerk & Umbgrove, 1927). therefore assume that the remains of Ngandong Von Koenigswald tried to apply this method to man are contemporaneous with the associated vertebrate collections, and he succeeded in esta­ fauna and with the High terrace formation of the blishing a vertebrate stratigraphy of Java, exten­ Solo river. ding from the Pliocene until into the Holocene. The differences between the successive faunas were explained by von Koenigswald in terms of migra­ 5. THE AGE OF THE NGA NDONG TERRACE tions from mainland Asia in combination with the isostatic movements of the sea level in the Pleisto­ A first attempt to find a relative date for the Solo cene, during which Java was either an island or part terraces and for the fossil remains therein, can be of the continent. made with a reconstruction of the history of the When the fauna of Ngandong was excavated, the Solo river drainage basin: in fact a reconstruction of fauna of Trinil was aiready well-known. Near this geomorphic progress in geologic time, a contro­ latter village early Homo erectus had been found at versial issue to say the least. Like other disciplines the end of the last century together with animal modern geomorphology has turned away from old remains. Von Koenigswald was able to show that evolutionary views, and the landscape is now stu­ this fauna of Trinil was considerably older than that died in functional, interrelated terms. Unchanged ofNgandong. Not only could he point to the higher geomorphic relics are viewed with suspicion. Yet percentage of recent species in the Ngandong fauna, the reconstruction of a drainage basin history is but he was also aware of the fact that the parent only possibie on the basis of information provided sediment of the Ngandong fossils was associated by relics like abandoned valley floors and fossil with the present Solo course, which could not be floodplalns. demonstrated at Trinil. Because von Koenigswald Lehmann (1936) has given an acceptable outline regarded the Trinil fauna as Middle Pleistocene, in of the history of the Solo drainage basin. This his view the Ngandong fauna was Upper Pleisto­ outline has never seriously been contested (van cene, and consequently this had also to be the age of Bemmelen, 1949; Sartono, 1976), and the results of Late Homo erectus or Ngandong man of Java 5

my own surveys only strengt hen Lehmann's views magnetic determinations that attribute a normal (Bartstra, 1984). It appears that the SQlo initially polarity to the Notopuro lahars (Semah, 1982; flowed southwards towards the Indian Ocean (via 1986). the now dry valley of Giritontro), and that not It has been argued that since part of the Notopuro earlier than the Upper Pleistocene a northward layers appears to be Middle Pleistocene, the Solo directed drainage system did develop, as a result of High terrace could still be Middle Pleistocene too epeirogenic movements'in South J ava. Consequent­ (Howell, 1986). A recently obtained U-series date of ly the oldest Solo terrace sediments in Central Java terrace bone fragments from the surroundings of must date from the beginning of this northward Ngandong, 165±i� ka, would appear to confirm this directed drainage pattern. view (Bartstra et al., 1988). But this date is the result Not toa long ago the Upper Pleistocene was still a of an analysis of a few small, rounded pieces of relative concept. But since geomagnetic reversals bone, that could be allochthonous in the High have been recorded in detail, it has been establis hed terrace. Vertebrate bone fragments from a small that the Upper Pleistocene is the period between the excavation in the still remaining High terrace end of the Brunhes normal, 125 ka ago, and the remnant at Ngandong itself gave much younger beginning of the Holocene, IO ka ago. The Middle dates, all within the Upper Pleistocene (table l: Pleistocene is congruent with the Brunhes normal, 31±ika/ 101±:� ka). Therefore, in my opinion the and spans the period from about 715 ka to 125 ka age of the Solo High terrace, and consequently that ago. It should thus be ascertained whether radio­ of Homo erectus soloensis, must still be taken to be metric determinations of Solo High terrace deposits Upper Pleistocene. or constituents still give good reason for designating U-series dates are more disputable than K-Ar cir these deposits as Upper Pleistocene. FT dates, because of still unknown factors that In the Pleistocene stratigraphy of Central Java influence the outcome of the analysis. So it is the Solo High terrace is superimposed on the so­ sometimes presumed that an outer surface or perio­ called Notopuro layers, volcanic tuff breccias or steal probe of vertebrate bone material is preferable lahars and associated stream-Iaid graveis, which to an inner bone analysis. Current research at the ' should be connected with the emergence of the Groningen Centre for Isotope Research may lead to Lawu volcano (Duyfjes, 1936; van Bemmelen, 1949; a different conclusion, however. As far as the Sartono, 1976). A glance at the map shows that the Ngandong ter race bones are concerned, the U­ Solo river partly encompasses this volcanic cone concentrations of the bulk are higher than those of (fig. 1). For the Notopuro deposits there exists one the surface (Bartstra, in press b; van der Plicht et al., radiometric FT (fission-track) date of 250±70 ka in press). (O.25±O.07 Ma), which comes, however, from pu­ mice pebbles from the fluvial part of these Noto­ puro sediments and thus is to be associated not with 6. THE ARTIFACTS OF NGANDONG MAN any direct lahar activity.but with subsequent river­ AND THE PALAEOLITHIC OF J AVA action (Suzuki et al., 1985), It stands to reason that these pebbles have been derived, and that the FT The excavations at Ngandong in the 1930's did not date marks the onset of the Lawu volcanism result in the find of in situ stone implements. But towards the end of the Middle Pleistocene. This mention was made of a so-called bone culture: view is not in disagreement with the few palaeo- pieces of antler, horns and bones with fracture

Table I. U-series ages from Ngandong bone samples (modified afte! Bartstra et al., 1988). 2 4 2J8 1JO 1 J4 1J lll 1J l ' Sample depth Code U-concen- 3 U/ U Th/ U OTh/ Th OTh/ 3 U age tration (ppm) (ka)

B (surface) G-86656 45 1.092±O.0Il 0.375±0.033 150±45 51±5 C(I.IOm) G-86657 12 1.I50±0.015 0,342±0.035 26±7 45±5 D (1.20 m) G-86658 18 1.I12±0,013 0.323±0,028 27±6 42±4 E (1.65 m) G-86659 82 1.20?±0.007 0,344±0.040 120±36 45±6 F (1.96 m) G-86688 128 1.417±0.005 0.409±0,058 56±lg G (2.20 m) G-86689 68 1.280±0.006 0.247±0.021 31±i

H (2.30 m) G-86690 54 1.252±0,006 0,540±0.040 118±30 82±7 J (2.32 m) G-86691 74 1.427±0.006 0.489±0.044 70#

K (2.50 m) G-87037 48 1.164±0.0 II 0.617±0.04l 93±31 101±:� 6 G.-J. BARTSTRA surfaces assumed to be artifactual (Oppenoorth, but a distinction between two tool-inventories is 1936; von Koenigswald, 1951). I have always had clear. If one assumes (with von Koenigwald's faunal my doubts about this presumed osteodontokeratic migrations in mind) that Homo sapiens came to Java culture from the Solo terraces, but it might be from the continent (Hooijer, 1950; 1952; Shutler, interesting to study the excavated bone collection 1984), he could have brought another tool-inven­ from Ngandong again, in view of current know­ tory to the island, in which the heavy core-types ledge with regard to anatomical part-frequencies or dominated. Gradually he may have explored the artificial marks. It would be very difficult, however, patches of trop icai forest and the caves of Java, to retrace any significalJt part of the original which his predecessor Homo erectus still avoided. collection. Surface finds of stone tools from the surroun­ dings of Ngandong were mentioned by von Koe­ 7. A FINAL REMARK nigswald (1933), who, during his field surveys in the Kendeng, was able to collect small chalcedony From the foregoing it emerges that much research artifacts from High terrace gravel. It stands to needs to be done in Java on the fascinating topic of reason that this surface gravel is the coarse residue late H01no erectus. From the Solo terraces strati­ of an originally much more varied terrace fill, and graphically secured samples of fossil vertebrate that these artifacts are in fact in situ. They are often material have to be collected in order to verify the abraded, as a result of fluvial transport. In the U-series dates from Ngandong. More good terrace literature they are referred to as: Ngandong culture, sections have to be found for the study of the fill-in­ Ngandong industry, or Ngandongian (Movius, fill phenomenon, and lithological probes of the 1949; van Heekeren, 1972; Soejono, 1975; Sartono, various gravels could help to elucidate High terrace 1980). divisions. Terrace outcrops with artifacts have to be As is the case with river-drift artifacts, they are mapped and investigated to underline the notion of fairly rare. One has to examine a lot of gravel before this basic distinction between an Early and a Late an artifact is found. The fact that surface gravel is Palaeolithic on Java. There is in fact so much to concerned makes the actual search in the field investigate in the fieids of Java, that before satis­ easier, but nevertheless the Ngandongian artifacts factory data can be provided, one needs many more found so far can be·spread out on the surface of an 'geo'-archaeologists. I think that satisfactory data average sized table. What then becomes apparent is can scarcely result from a short visit to some locality the absence of the classic, so-called Far Eastern by a single researcher. On Java one needs a well­ heavy core-types, as defined by Movius (1949): the coordinated team of interdisciplinary workers who choppers, the hand-adzes and the hand-axes. In­ can spend adequate time in the fieid. This is a stead, one notices in particular small flakes, rarely challenge that has to be met, in the first place by the exceeding 5 cm in length, and only occasionally institutes directly involved: the Pusat Penelitian transforrned into true implements, like scrapers or Arkeologi Nasional in Jakarta, and the Biologisch­ gravers. The rare cores and core-tools are also Archaeologisch Instituut in Groningen. small-sized and are very different from Movius' types. I presurne that these Ngandongian artifacts can indeed be associated with Homo erectus soloen­ 8. ACKNOWLEDGEMENT sis or Ngandong man (Bartstra et al., 1988). There exists also a heavy chopper chopping-tool I wish to thank Dr. Richard Shutler Jr. for his industry on Java, the so-called Pacitan industry, critical comments on the first draft of this paper; concentrated in the karstlriverine landscapes along and Dr. Jan Wisseman for the lively discussion in Java's south coast. Although this industry can still front of the outcrop of Grogol. be called Palaeolithic, it is in my opinion ve ry young, and it dates surely from the second half of the Upper Pleistocene (Bartstra, 1984). This Paci­ 9. REFERENCES tanian should be associated with Homo sapiens (Wajak man?) who inhabited Javå at that time. So one cannot avoid the impression that on Java BARTSTRA, G.-J., 1977. The height of the river terraces in the transverse Solo valley jn Java. Modem Qualernary Research there is a basic distinction between what I would call in SOl/lheasl Asia 3, pp. 143-155. the Early Palaeolithic (the Solo High terrace (Homo BARTSTRA, G.-J., 1982. Homo erecll/s ereclus: the search for erecfus) Palaeolithic) and the Late Palaeolithic (the his artifacts. Currenl Anlhropology 23, pp. 318-320. Pacitanian chopper/chopping-tool (Homo sapiens) BARTSTRA, G.-J., 1984. Dating the Pacitanian: some thoughts. Courier Forschungs-!nslilul Senckenberg 69, pp. 253-258. Palaeolithic; Bartstra & Basoeki, in press). Un­ BARTSTRA, G.-J., in press a. Indonesia during the period of doubtedly local differences play a role as a con­ Homo sapiens neanderlhalensis. In: Hislory of Ihe scienlijic sequence of the former availability of raw material, and cullural developll1enl ofmankind (UNESCO). Paris; Late Homo erectus or Ngandong man of Java 7

BARTSTRA, G.-J., in press b. The age of Solo man and Java LEHMANN, H., 1936. Morphologische Studien auf lava. Stutt­ man. Archaeome/ry: Aus/ralasian Studies 2. gart. BARTSTRA, G.-J. & BASOEKI, in press. Recent work on the MARTIN, K., 1919. Unsere palaeozoologische Kenn/nis von lava. Pleistocene and the Palaeolithic of Java. Curren/ An/hropo­ Leiden. logy. MOVIUS, H.L., 1949. The Lower Palaeoli/hic Cl/I/I/res of sou/hem

BARTSTRA, G.-J., S. SOEGONDHO & A. VAN DER WIJK, and eas/em Asia ( = Trans. Arner. Phil. Soc. n.s. 38). 1988. Ngandong man: age and artifacts. loumal of HI/lilan Cambridge (Mass.). Evolu/ion 17, pp. ,325-337: OPPENOORTH, W.F.F., 1932a. HOlllo (lavan/hropus) soloen­ BEMMELEN, R.W. VAN, 1949. The geology of Indonesia. The sis. Een plistoceene mensch van Java. Voorlopige mede­ Hague. deeling. We/enschappelijke Mededelingen Diens/ van de Mijn­ BINFORD, L.R. & N.M. STONE, 1986. Zhoukoudian: a closer bouw Nederlands-Indie 20, pp. 49-75. look. Cunen/ An/hropology 27, pp. 453-468. OPPENOORTH, W.F.F., 1932b. Ein neuer diluvialer Urmensch DA Y, M.H., 1986. GI/ide /0 fossil man. 4th ed. London. von Java. Na/ur und MusellIn 62, pp. 269-279. DUYFJES, J., 1936. Zur Geologie und Stratigraphie des Ken­ OPPENOORTH, W.F.F., 1936. Een prehistorisch cultuur­ denggebietes zwischen Trinil und Soerabaja (Java). De centrum langs de Solo-rivier. Tijdschrif/ van hel Koninklijk Ingeniel/r in Ned.-Indie, Mijnbol/w en Geologie, de Mijn­ Nederlands-Aardrijkskundig Genoo/schap 53, pp. 399-411. ingenieur 3, pp. 136-149. PLICHT, J. VAN DER, A. VAN DER WIJK & G.-J. BART­ ELBERT, J., 1908. Ueber das Alter der Kendeng-Schichten mit STRA, in press. Uranium and thorium in fossil bones: Pi/hecan/hropus erec/us Dubois. Neues lahrbl/ch fur Minera­ activity ratios and dating. Applied Geochemis/ry 346. logie, Geologie und Paliion/ologie 25, pp. 648-662. POPE, G.G.,.1985. Taxonomy, dating and paleoenvironment: HAAR, C. TER, 1934. Homo-soloensis. De Ingenieur in Ned.­ the paleoecology of the early Far Eastern hominids. Modem Indie, Mijnbouw en Geologie, de Mijningenieur I, pp. 51-57. QI/a/ernary Research in Sou/heas/ Asia 9, pp. 65-80. HARDJASASMITA, H.S. & D. MULYANA, 1986. Beberapa PUECH, P.F., 1983. Tooth wear, diet, and the artifacts of Java sisa keaktifan manusia dimasa lampau di Gua Lowo Sam­ man. Cunelll An/hropology 24, pp. 381-382.

pung. Per/eml/an I/miah Arkeologi 4, pp. 109-124. SANTA LUCA, A.P., 1980. The Ngandong fossil 1I0minids (= HEEKEREN, H.R. VAN, 1972. TheS/oneAge oflndonesia. 2nd Yale Univ. Publ. in Anthropology 78). New Haven.

rev. ed. ( = Verhandelingen Kon. Inst. voor Taal-, Land- en SARTONO, S., 1976. Genesis of the Solo terraces. Modem Volkenkunde 61). The Hague. Qua/emU/y Research in Soll/heas/ Asia 2, pp. 1-21. HOOIJER, D.A., 1948a. Prehistoric teeth of man and of the SARTONO, S., 1980. Homo erec/us ngandongensis: the possibIe orang-utan from Central Sumatra, with notes on the fossil maker of the 'Sangiran tlakes'. An/hropologie 18, pp. 121-131. orang-utan from Java and southern China. Zoologische SEMAH, F., 1982. Pliocene and Pleistocene geomagnetic re­ Mededeelingen 29, pp. 175-301. versals recorded in the Cemolbng and Sangiran dornes HOOIJER, D.A., 1948b. Evolution of the dentition of the orang­ (Central Java). Modem Qua/emary Research in Southeas/ utan. Na/ure 162, p. 306. Asia 7, pp. 151-164. HOOIJER, D.A., 1949. Mammalian evolution in the Quater­ SEMAH, F., 1986. Le peuplement ancient de Java. Ebauche nary of southern and eastern Asia. Evolu/ion 3, pp. 125-128. d'un cadre chronologique. I'An/hropologie 90, pp. 359-400. HOOIJER, D.A., 1950. Fossil evidence of Austromelanesian SHUTLER, R., 1984. The emergence of Homo sapiens in migrations in Malaysia? Sou/lllves/em loumal of An/hropo­ Southeast Asia, and other aspects of hominid evolution in logy 6, pp. 416-422. East Asia. In: R.O. Whyte (ed.) The evolu/ion of /heEas/ Asian HOOIJER, D.A., 1952. Austromelanesian migration once more. evironlllen/, 2. Hongkong, pp. 818-821. Sou/hwes/em lOl/mal of An/hropology 8, pp. 472-477. SOEJONO, R.P., 1975. SejU/'ah Nasional Indonesia I. Jakarta. HOWELL, F.C., 1986. Variabilite chez Homo erec/us, et pro­ SUZUKl, M., WIKARNO, BUDlSANTOSO, J. SAEFUDIN & bIernede la presence de cette espece en Europe. l. Appendice. M. ITIHARA, 1985. Fission track ages of pumice tuff, tuff I'An/hropologie 90, pp. 465-481. layers and javites of hornin id fossil bearing formations in JACOB, T., 1972. The problem of head-hunting and brain­ Sangiran area, Central Java. In: N. Watanabe & D. Kadar eating among Pleistocene men in Indonesia.Archaeology and (eds.), Qua/emary geology of /lIe Iwminid fossil bearing Physical An/hropology in Oceania 7, pp. 81-91. formations in lava. Bandung, pp. 309-331. JACOB, T., 1975. Indonesia. In: K.P. Oakley, B.G. Campbell & TEILHARD DE CHARDlN, P., 1956. I'Appari/ion de I' holllme. T.1. Molleson (eds.), Ca/alogue of fossil hOlllinids. London, Paris. pp. 103-116. TERRA, H. DE, 1943. Pleistocene geology and early man in JACOB, T., 1978. The puzzle of Solo man. Modem Qua/emary Java. Transac/ions American Pleis/ocene Socie/y n.s. 32, pp. Research in Sou/heas/ Asia 4, pp. 31-40. 437-464. KOENIGSWALD, G.H.R. VON, 1933. Rapport over een VLERK, J.M. VAN DER & J.H.F. UMBGROVE, 1927.

controlebezoek aan Ngandong. Unpubl. report former Ter/iaire gidsforaminiferen van Nederlandsch Oos/ Indie ( = Dutch-Eastindies Geol. Survey. Wetensch. Meded. Dienst v.d. Mijnb. Ned. Indi.e 6). Ban­ KOENIGSWALD, G.H.R. VON, 1951. Introduction. In: F. dung.

Weidenreich, Morphologie of Solo lilan (= Anthrop. Papers VOS, J. DE, 1985. Faunal stratigraphy and correlation of the Arner. Museum of Natural History 43). New York, pp. Indonesian hominid sites. In: E. Delson (ed.), Allces/ol'S.' /lIe 211-221. hard evidence. New York, pp. 215-220. KOENIGSWALD, G.H.R. VON, 1955. Begegnungen. mi/ dem WEIDENRETCH, F., 1945. Gian/ early man from lava andSOI//h Vormenschen. DiisseldorflKiiln. China (= Anthrop. Papers Arner. Museum of Natural KOENIGSWALD, G.H.R. VON, 1958. Der Solo-Mensch von. History 40, I). New York. Java: ein tropischer NeanderthaIer. In: G.H.R. von Koe­ WEIDENREICH, F., 1951. Morphology of Solo man (= An­ nigswald (ed.), Hunder/ lahre Neander/haler. Utrecht, pp. throp. Papers Amer. Museum of Natural History 40,3). New 21-26. York.